Update Dockerfile

Dockerfile

@@ -1,37 +1,27 @@
+# Use a small Python base
 FROM python:3.11-slim
 
-# System deps for llama-cpp build
+# System deps (curl only for quick debugging if needed)
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    build-essential cmake git && \
-    rm -rf /var/lib/apt/lists/*
+    curl \
+ && rm -rf /var/lib/apt/lists/*
 
+# Set persistent cache location for Docker Spaces
+ENV HF_HOME=/data/hf_cache
+ENV HUGGINGFACE_HUB_CACHE=/data/hf_cache
+
+# App dir
 WORKDIR /app
 
-# App deps
-COPY requirements.txt .
+# Dependencies
+COPY requirements.txt /app/requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt
 
-# --- Bake model into the image (change these ENV if you prefer Phi-3 mini) ---
-ENV MODEL_REPO=Qwen/Qwen2.5-3B-Instruct-GGUF
-ENV MODEL_FILE=qwen2.5-3b-instruct-q4_k_m.gguf
-
-# Download GGUF at build time to /app/models (no symlinks)
-RUN python - <<'PY'
-from huggingface_hub import snapshot_download
-import os
-local_dir = "/app/models"
-os.makedirs(local_dir, exist_ok=True)
-snapshot_download(
-    repo_id=os.environ.get("MODEL_REPO"),
-    allow_patterns=[os.environ.get("MODEL_FILE")],
-    local_dir=local_dir,
-    local_dir_use_symlinks=False,
-)
-print("Model baked into image at /app/models")
-PY
-
 # App code
-COPY app.py .
+COPY app.py /app/app.py
 
+# Expose Space port
 EXPOSE 7860
+
+# Start the API
 CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]